Three additional key considerations which affect extended-range DLVA units are discussed in this
section:
a) the crossover match, i.e. the behavior of the transfer curve when crossing from the
region where the first detector is the active signal source to the region where the second
detector is the active signal source;
b) the limiting behavior of the RF amplifier; and
c)
the optimum power level at which to start logging.
At a fixed temperature and RF frequency, i.e. at constant RF gain, the roll-off of the first detector
(and LVA), and the onset of the second DLVA have to be matched to ensure a smooth transition.
This can be accomplished by adjustment of the linear extension for DLVA #1 and the L1 current
for DLVA #2.
It is important to realize that the problem referred to in section 8.2 (and illustrated in Fig. 8.2),
regarding the divergence, separation, and re-convergence of the log transfer curves at different
temperatures (which is caused by the temperature-dependent gain of the log stages) does not re-
occur in the crossover region of a double-DLVA. This is for the same reason that it is not a further
problem after the first log stage in a single DLVA unit. The output of the second DLVA's first log
stage will limit at a lower input power at low temperature than it will at a higher temperature. Thus
the differential onset of logging at different temperatures is exactly offset.
However, if the gain of the RF amplifier depends on both temperature (T) and the RF frequency
(f), the shape of the log transfer curve will be a function of both f and T. Nothing can be done in
the video section to compensate the frequency dependence of the transfer curve, but if the RF
amplifier gain G(f,T) can be factored as follows,
G ( f , T ) = G1(f) x G2(T)
then the temperature-dependence of the transfer curve can be fixed in the video section.
To get good crossover behavior of the transfer curve, the initial limiting should come from the L-
17C, and not from the RF section. However, the RF section should limit as soon as possible after
the L-17C has, in order to facilitate trimming for good recovery times, as discussed further below.
Concerning the question of where to set the start of logging, dual DLVAs differ from single units in
a very important way. Earlier, in discussing the case of single DLVA units, we showed that with
respect to both recovery time and temperature behavior, the power-level at which logging is
initiated is not critical, but merely cosmetic, if recovery and temperature behavior are only
specified in the logging range. In the case of a dual or triple unit, this is not true.
Consider the case of temperature-induced shifts for a double-DLVA unit. When the low-power unit
(#1) is active (i.e. into logging), the high-power unit (#2) may be seeing negligible power.
-32-
 |
 |
 |
 |
 |